Tools for reduced precision computation: a survey
The use of reduced precision to improve performance metrics such as computation latency
and power consumption is a common practice in the embedded systems field. This practice …
and power consumption is a common practice in the embedded systems field. This practice …
Sirnn: A math library for secure rnn inference
Complex machine learning (ML) inference algorithms like recurrent neural networks (RNNs)
use standard functions from math libraries like exponentiation, sigmoid, tanh, and reciprocal …
use standard functions from math libraries like exponentiation, sigmoid, tanh, and reciprocal …
Energy-efficient approximate multiplication for digital signal processing and classification applications
The need to support various digital signal processing (DSP) and classification applications
on energy-constrained devices has steadily grown. Such applications often extensively …
on energy-constrained devices has steadily grown. Such applications often extensively …
Compiling KB-sized machine learning models to tiny IoT devices
Recent advances in machine learning (ML) have produced KiloByte-size models that can
directly run on constrained IoT devices. This approach avoids expensive communication …
directly run on constrained IoT devices. This approach avoids expensive communication …
An introduction to distributed smart cameras
Distributed smart cameras (DSCs) are real-time distributed embedded systems that perform
computer vision using multiple cameras. This new approach has emerged thanks to a …
computer vision using multiple cameras. This new approach has emerged thanks to a …
Real-time video analysis on an embedded smart camera for traffic surveillance
M Bramberger, J Brunner, B Rinner… - … . RTAS 2004. 10th …, 2004 - ieeexplore.ieee.org
A smart camera combines video sensing, high-level video processing and communication
within a single embedded device. Such cameras are key components in novel surveillance …
within a single embedded device. Such cameras are key components in novel surveillance …
A holistic approach to automatic mixed-precision code generation and tuning for affine programs
J Xu, G Song, B Zhou, F Li, J Hao, J Zhao - Proceedings of the 29th ACM …, 2024 - dl.acm.org
Reducing floating-point (FP) precision is used to trade the quality degradation of a numerical
program's output for performance, but this optimization coincides with type casting, whose …
program's output for performance, but this optimization coincides with type casting, whose …
[PDF][PDF] A Programmable Architecture for Real-Time Derivative Trading
S Tandon - Master's Thesis, University of Edinburgh, 2003 - olsendata.com
Derivatives are financial securities that are used to hedge business risks, caused by
changes in foreign exchange rates, interest rates or prices of goods. The algorithms in the …
changes in foreign exchange rates, interest rates or prices of goods. The algorithms in the …
Shiftry: RNN inference in 2kb of RAM
Traditionally, IoT devices send collected sensor data to an intelligent cloud where machine
learning (ML) inference happens. However, this course is rapidly changing and there is a …
learning (ML) inference happens. However, this course is rapidly changing and there is a …
Application of symbolic computer algebra in high-level data-flow synthesis
A Peymandoust, G De Micheli - IEEE transactions on computer …, 2003 - ieeexplore.ieee.org
The growing market of multimedia applications has required the development of complex
application-specified integrated circuits with significant data-path portions. Unfortunately …
application-specified integrated circuits with significant data-path portions. Unfortunately …